Proposed is a demonstration and implementation project to increase the appropriate use of aspirin for primary prevention of acute myocardial infarction and stroke in the population of Minnesota. Although significant progress in the reduction of acute myocardial infarction and stroke is apparent, these cardiovascular disorders continue as the leading causes of morbidity and mortality. In recent years, and in the context of the positive results from large randomized clinical trials, there is growing consensus that aspirin when appropriately used, reduces cardiovascular morbidity in men aged 45-79 and women aged 55-79 years. The use of aspirin for primary prevention was recommended in the U.S. Preventive Services Task Force Report in 2009 and more recently in the CDC/CMS 'Million Hearts' ABCs (aspirin, blood pressure, cholesterol, smoking) campaign. Aspirin use is also recommended in the Healthy People 2020 goals. However, many adults who would benefit are not taking aspirin. We have been fortunate to receive a large philanthropic donation to launch a campaign to increase appropriate aspirin use in the State of Minnesota. The proposed grant requests funding to evaluate that campaign and the innovative approaches it proposes. In the context of an innovative mass media effort, we intend to test new methods for health system change to increase appropriate aspirin use. Using a two-arm design, we have defined 24 geographic areas in the state that will form the basis for a group-randomized trial. This design will allow us to distinguish the effects of the intervention methods. The interventions will be evaluated by sequential surveys of the target-age general population at baseline and follow-up. Appropriate aspirin use in that population will be the primary goal and endpoint. Simultaneous surveys of adjacent Upper Midwestern states (Iowa, North and South Dakota, Wisconsin) will assess secular trends. Morbidity and mortality data will monitor disease trends and complications associated with aspirin use. A substantial pilot study in a middle-sized community in Northern Minnesota allows us to refine and validate our intervention and measurement methods at the community level. This pilot also demonstrated behavior change of a magnitude supportive of our design assumptions. This program, if successful, should result in a generalizable program tested in a real world population setting.